Advanced ceramics and ceramic composites have a variety of advantageous properties, such as high hardness, strength, and wear resistance. This makes them good candidates for materials in the aerospace, automotive and defense industries, among others. A major disadvantage of advanced ceramics and ceramic composites is their requirement for specialized processing which often makes manufacturing complex shaped ceramic objects challenging and costly. Additionally, these materials are susceptible to flaw incorporation during production. These flaws are initiation points for failure and thus lead to a drastic reduction in strength. Repeatable manufacturing methods and optimized processes are compulsory for cost saving and production of high quality parts. In recent years, new processing technologies, such as gelcasting, have been developed to accommodate the formation of complex shaped ceramics and also the manufacture of ceramic composites. The use of wet forming technologies, like slip casting or gelcasting, necessitates the careful drying of ceramic objects. Complex shaped objects are particularly difficult to properly dry without introducing internal stresses which may result in warping and cracking, thus rendering the object unusable. Additionally, traditional drying processes are often energy intensive and lengthy, neither of which are favorable in a production setting. To improve manufacturability, the processing-structure-property relationships developed during the drying process must be investigated further. This work addresses the need to define optimized process conditions for the drying of alumina objects gelcast using gelatin. The osmotic drying process was employed to remove solvent from the objects through the use of an aqueous liquid desiccant solution of poly(ethylene glycol) (PEG). The process settings for the solution’s osmotic pressure and molecular weight were investigated, in addition to the total immersion time. The mass transfer processes that occurred between the ceramic object and the liquid desiccant solution were quantified in several case studies. For one sample, 40 weight% of the initial water content was removed in 75 minutes demonstrating the potential drying efficiency of this method. Depending on the initial solution conditions, the PEG solute was found to diffuse into the ceramic object to varying degrees. The effect of the drying condition on the object’s density and hardness was also measured. Through the development of regression equations, the process settings were optimized based on the goals to maximize water loss, minimize solids gain, and maximize the object’s density. The optimum drying settings for the objects studied in this work were an osmotic pressure of 2.50 MPa, a molecular weight of at least 100,000 g/mol, and an immersion time of 60 minutes. When objects of similar geometry, composition, and solution-to-object volume ratio are immersed in this type of solution, they are expected to lose 28 weight% of the object’s initial water content, gain solids of 0.82 weight% of the object’s initial mass, and have a density of 3.54 g/cm3. Furthermore, the regression models were validated using an independent experimental study. A model based on mass balance was used to define the kinetics of the mass transfer, along with the equilibrium values. Lastly, a demonstration of the feasibility of combining gelcasting, osmotic drying, and sacrificial templating is presented. Overall, these results may be used as the basis for further investigation into the scale up of the osmotic drying of gelcast alumina with the eventual implementation of the process in an industrial setting. / A Dissertation submitted to the Department of Industrial and Manufacturing Engineering in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Summer Semester 2018. / June 27, 2018. / alumina, drying, gelcasting, liquid desiccant, osmotic drying, preform / Includes bibliographical references. / Okenwa Okoli, Professor Directing Dissertation; Simone Peterson Hruda, University Representative; Zhiyong Richard Liang, Committee Member; Mei Zhang, Committee Member.

This study embarked with focus on the analysis, to quantify the potential impacts of varying market penetration rates (MPR) of connected and autonomous vehicles (CAVs) on the traffic network mobility and safety in construction work zones. The prospects for success were evaluated in three ways: (1) understanding the potential benefits of CAV in improving the travel time as result of work zone impacts; (2) reducing the delay and queues by improved time headway; and (3) safety benefits of the autonomous vehicles at varying market penetration rates by lowering the levels of incident probabilities and the Time-to-Collision. CAV market penetration rates, from 10% to 100%, in an increment of 10%, were used in the assessment of the potentials for these vehicles to improve the traffic performance in work zones. The effects at low and high penetration rates were assessed to evaluate the improvements on traffic mobility and overall network safety in the work zone when compared with traffic stream of only conventional vehicles. The motivation of this research comes primarily from the consideration of deployment and full utilization of connected autonomous vehicles in the real traffic world, most especially in the work zones which typically require altered road geometry such as lane closure and reduced speed areas. With varied market penetration rates (relative composition in the traffic) of the connected autonomous vehicles, the research question is premise on the actualization and an assessment of the safety benefits and improvement of network performance. VISSIM, a microscopic traffic simulation software, was used extensively in the analysis of the traffic conditions because it is able to represent different driver behaviors and interaction between vehicles. A broad literature review in the area of microsimulation of connected and autonomous vehicles was conducted to examine the effects of automated vehicles effects on traffic operations. Two different work zone locations were considered for this study: one is a long stretch of work zone on Interstate 44 in Missouri; and the other is a short stretch on Interstate 95 in Jacksonville, Florida. The study sites do not have similar traffic characteristics and geometry and hence the impacts of CAVs on both sites were evaluated independently. Different driver behavior models comprising cautious, moderate and assertive types, were considered for the study. After a careful analysis of the impacts of the different levels of CAV capabilities on the two work zones, findings show that CAVs have potential benefits of improving the traffic situations in work zones. For instance it was observed that even at low MPR of 10% CAVs, travel time was reduced by 14%. A monotonic improvement in travel time was observed across all CAV market penetrations. Results also showed that the benefits of travel CAVs were only noticeable when there is a high demand in traffic or during the peak period. No noticeable changes in traffic performance were observed when demand was low. Similar results were observed for impacts that CAVs have on both delays and queue length; this is because both of the measures of performance are related. When CAVs assumed the most cautious driver behavior models, results showed that the average vehicle delay was reduced, and there was a gradual improvement in delay as MPR increases. When market penetration approaches 100%, the work zone’s average vehicle delay could be reduced by a 70%. For the moderate (between cautious and aggressive) models, the average vehicle delay was greatly reduced. As CAV market penetrations approached 20%, similar results were achieved for the two scenarios of the moderate models that were evaluated. The results of the aggressive models appears to be more unrealistic, and reduction in delays very dramatic at lower market penetrations of CAVS, resulting in an 85% reduction in average vehicle delay as the CAV fleet reached 20%. A sensitivity analysis of the VISSIM driver behavior parameters will be necessary in this case to assess the impacts of such parameters on the traffic flow. In general, introduction of CAVs in the traffic stream lowered the average vehicle delays, however the resulting impacts vary across the different driver behavior models. The impacts on CAVs was also assessed for safety of the road users; the time-to-collision (TTC) and post encroachment time (PET) surrogate safety measures were employed to determine the effects of CAVs on the presence of conflicts. Findings from this study shows that as CAV market penetrations increase, the conflicts in the work zones reduced significantly. CAVs with more assertive driver behaviors indicated a higher percentage in conflict reductions even at lower market penetrations. Though the impacts vary for all the behavior models, it was observed that there is capability-wide safety benefits for CAV market penetrations. However, the proportion of conflicts in the severity zones increased with an increase in the CAV aggressiveness. / A Thesis submitted to the Department of Civil and Environmental Engineering in partial fulfillment of the requirements for the degree of Master of Science. / Summer Semester 2018. / July 19, 2018. / Includes bibliographical references. / John Sobanjo, Professor Directing Thesis; Yassir AbdelRazig, Committee Member; Eren Erman Ozguven, Committee Member.

Object detection is an important part of the image processing system, especially for applications like Face detection, Visual search engine, counting and Aerial Image analysis. Hence the performance of object detectors plays an important role in the functioning of such systems. With the advancements in the Deep learning field, Convolutional Neural Networks (CNN) is now the state of the art in object detection and classification. In this thesis, we have investigated different object detection models and studied the chain of developments which has taken place from one model to the next, the improvements over the previous models are studied by comparing the model architecture, methods of extracting features and by the performance on different object detection and classification competitions. The thesis is divided into three parts, first, a few notable models developed before the deep learning era are studied, these models have hand-coded feature extractors. In the next part of the thesis, Convolutional Neural Networks, part of deep learning for object detection has been researched by understanding the basic Convolutional Neural Network architecture and then different ways to improve it. It is seen that the CNN object detectors use a base architecture, these architectures can be classified as basic and advanced. In the final part of the thesis, the Meta architectures, which are divided as region based and regression based models are presented in detail, with depictions of such implementations. These Meta architectures are based on the basic and advanced architectures studied in the previous sections. With a look at the hand coded features based object detectors in the beginning, this work ends with a comparison of the state-of-the-art models discussed in the Meta architecture section. In the end, the future scope of the object detectors and possible new applications are discussed. Keywords: Neural Networks, Convolutional Neural Networks, object detectors, Deep learning, regression based, region based / A Thesis submitted to the Department of Electrical and Computer Engineering in partial fulfillment of the requirements for the degree of Master of Science. / Summer Semester 2018. / July 16, 2018. / Convolutional Neural Networks, Deep learning, Neural Networks, object detectors, region based, regression based / Includes bibliographical references. / Simon Foo, Professor Directing Thesis; Bruce A. Harvey, Committee Member; Pedro Moss, Committee Member.

Aerodynamics of road vehicles have continued to be a topic of interest due the relationship between fuel efficiency and the environmental impact of passenger vehicles. With the streamlining of ground vehicles combined with years of geometric and shape optimization, other techniques are required to continue to improve upon fuel consumption. One such technique leverages aerodynamics to minimize drag through the implementation of flow control techniques. The current study focuses on the application of active flow control in ground vehicle applications, employing linear arrays of discrete microjets on the rear of a 25 Ahmed model. The locations of the arrays are selected to test the effectiveness of microjet control at directly manipulating the various features found in typical flow fields generated by ground vehicles. Parametric sweeps are conducted to investigate the flow response as a function of jet velocity, momentum, and vehicle scaling. The effect and effciency of the control are quantified through aerodynamic force measurements, while local modifications are investigated via particle image velocimetry and static pressure measurements on the rear surfaces of the model. Microjets proved most effective when utilized for separation control producing a maximum change to the coefficients of drag and lift of -14.0% and -42% of the baseline values, respectively. Control techniques targeting other flow structures such as the C-pillar vortices and trailing wake proved less effective, producing a maximum reduction in drag and lift of -1.2% and -7%. The change in the surface pressure distribution reveals the impact of each flow control strategy on a targeted flow structure, and highlights the complex interaction between the salient flow features found in the wake of the Ahmed model. Areas of pressure recovery on the surface of the model observed for each control technique support the observed changes to the aerodynamic forces. The time averaged, volumetric wake is also reconstructed to characterize the baseline flow field and highlight the effect of control on the three dimensional structure of the near wake region. The results show that separation control has a measurable effect on the flow field including modifications of the locations, size, magnitude, and trajectory of the various structures which comprise the near wake. The observations give insight into desirable modifications and flow topology which lead to an optimal drag configuration for a particular vehicle geometry. / A Dissertation submitted to the Department of Mechanical Engineering in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Summer Semester 2018. / July 13, 2018. / active flow control, aerodynamics, ahmed model, drag reduction / Includes bibliographical references. / Farrukh Alvi, Professor Directing Dissertation; Sungmoon Jung, University Representative; Rajan Kumar, Committee Member; Kunihiko Taira, Committee Member; Seungyong Hahn, Committee Member.

One of the major concerns for many athletes in todays' sports is mild Traumatic Brain Injury (mTBI), which is commonly known as concussion. Researchers and manufacturers of sport helmets are constantly trying to develop new designs and technologies to better prevent mTBI. The objective of this research is to study the effective designs for sport helmets that can potentially absorb and dispel both linear and rotational forces acting on the head during impact. Inspiration by the different types of working mechanisms and structures existing in nature that can absorb energy from different types of impacts, new designs were explored. Honeycomb structures have been used extensively in lightweight sandwich structure and impact energy absorption applications. Recently, Auxetic structures are attractive for various engineering applications because of their unique mechanical properties, volume change control and excellent impact energy absorption performance. In this study, novel designs and performance improvement of new auxetic-strut structures were presented. A comparative study of in-plane and out-of-plane uniaxial compression loading behavior of regular honeycomb, re-entrant auxetic honeycomb, locally reinforced auxetic-strut structure and a hybrid structure of combining regular honeycomb and auxetic-strut structure was conducted. Finite element modelling was carried out to reveal their structure-property relationships. The deformation and failure modes of the different designs were studied and their performance was also discussed. The new auxetic-strut structure showed better mechanical properties than the honeycomb and auxetic structures with a small density increase. For in-plane performance, the compressive strength of the auxetic-strut design is ~300% more than that of honeycomb structure and ~65% more than that of auxetic structure. With lower values of the Poisson’s ratio, the new design can absorb more energy when compared to the other structures. The out-of-plane properties of auxetic-strut design showed an increase of ~68% in the compressive strength, ~63% in Young’s modulus and ~32% in the total energy absorbed when compared with the honeycomb structure. The hybrid structures also showed excellent out-of-plane properties. With better in-plane and out-of-plane properties, auxetic-strut design can be used in various energy absorption applications. Hybrid designs allow us to tailor properties of the structures with their specific in-plane and out-of-plane deformation and failure modes A comparative study of dynamic crushing behavior of the structures was also carried out. Finite element modelling was conducted to compare the dynamic crushing behavior of these structures at different impact velocities. Deformation mechanisms of these structures were studied, that provided the new insights on how to control the deformation of the structure and tailor the properties. For in-plane impact tests the energy absorbed by auxetic- strut and hybrid structures was half when compared with honeycomb and re-entrant auxetic structures at lower strain levels. But at higher strain levels, the new structures performed twice as that of the later. In contrary for out-of-plane crushing, the energy absorbed by the auxetic-strut and hybrid structures was higher than the honeycomb and re-entrant auxetic structures at lower strain levels and vice-versa. Advanced manufacturing or 3D printing method were employed to produce samples of the new designs. The results of the sample tests are in good agreement with the modeling predictions. These results are valuable to provide new fundamental understanding of structure-property relationships for new auxetic-strut and their hybrid honeycomb structures for potential aerospace and sporting product applications, especially in football helmets. / A Dissertation submitted to the Department of Industrial and Manufacturing Engineering in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Spring Semester 2018. / February 08, 2018. / Auxetic-honeycomb hybrid structures, Auxetic structures, Energy absorption, Helmets, Honeycomb, In-plane, out-of-plane properties / Includes bibliographical references. / Zhiyong (Richard) Liang, Professor Directing Dissertation; Sungmoon Jung, University Representative; Changchun (Chad) Zeng, Committee Member; Tarik Dickens, Committee Member.

In recent years there has been a large push for magnet technologies extending into the 20 to 30 T range. High field solenoids can be made using resistive materials like Cu, however the resistive losses associated with Cu electromagnets is large (~ 20 MW for 31 T Cu solenoid). Superconductors, when cooled below a critical temperature, conduct electricity without resistive losses, significantly reducing the energy needed to maintain a magnetic field. For this reason, superconductors are used extensively in magnet technology. For the last 50 years, technologies have been dominated by the low temperature superconductors (LTS): Nb3Sn and Nb-Ti, which are metallic, isotropic, multi-filamentary, twisted, round wire. Despite their wide use, LTS conductors are limited by their irreversibility fields to below ~ 24 T (@ 4.2 K). If higher fields are to be reached, other materials must be used. The cuprate high temperature superconductors (HTS) have irreversibility fields 3 to 4 times larger than those of LTS conductors, making them a promising technology in the pursuit of superconducting magnets in exceeding 24 T. Limitations exist for the cuprate conductors, namely their electrical anisotropy, requiring the material be textured if large current densities are to be maintained. HTS conductors like YBCO and Bi-2223 achieve this texture by fabrication as anisotropic flat tapes. However, Bi-2212 is special in that it is the only HTS material capable of achieving texture in a round wire form with a metallic matrix used in the more common LTS technologies. Furthermore, Bi-2212 is multi-filamentary and can be twisted, while also being macroscopically isotropic with respect to magnetic field. For these reasons, Bi-2212 stands alone as the only HTS with the ability to use the most common cabling technologies originally developed for LTS conductors. The limitation of Bi-2212 is its brittle ceramic filaments, which must be well protected from significant amounts of strain, especially since 2014, when Bi-2212 experienced a large jump in critical current density upon using overpressure heat treatment (OPHT). We report here on ways of reinforcing Bi-2212 on the strand and coil levels. We show through experiment and modeling that, with a combination of reinforcing methods, Bi-2212 has the potential to reach high magnetic fields especially when used as insert coils. In preparation for use at high fields, we also present here high field measurements up to 31 T, and address the cause of large variation seen in the JC for Bi-2212 wires produced over the last decade. We attribute these large differences to variations in the connectivity between grains, altering the effective cross-section of the conductor. Lastly, we introduce a fit for determining the high field behavior (up to 31 T) of Bi-2212 wires from low field data (< 15 T) that is more easily attainable in common laboratories which do not have immediate access to fields in excess of 20 T. / A Dissertation submitted to the Department of Mechanical Engineering in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Fall Semester 2018. / November 16, 2018. / Includes bibliographical references. / David C. Larbalestier, Professor Directing Dissertation; Simon Foo, University Representative; William Oates, Committee Member; Sastry Pamidi, Committee Member; Seungyong Hahn, Committee Member.

Halide perovskites bring an unprecedented opportunity for low-cost high performance optoelectronic devices due to their extraordinary optical and electrical properties along with their solution processible nature. The record power conversion efficiency (PCE) of perovskite solar cells (23.3%) has surpassed polycrystalline silicon, copper indium gallium selenide (CIGS), and cadmium telluride (CdTe). In addition, the record external quantum efficiency (EQE) of perovskite light-emitting diodes (20.3%) is on par with the organic light-emitting diodes (OLEDs) and quantum dot light-emitting diodes (QLEDs). Benefiting from the superb properties of perovskites, there is increasing interests in fabricating perovskite optoelectronics in a large scale as well as with low elastic modulus (high flexibility and stretchability). Although the efficiencies of perovskite optoelectronics increase dramatically in the past few years, there are still concerns that cause short lifespan in perovskite optoelectronics, such as ion migration induced intrinsic perovskite instability, oxygen, moisture, non-radiative recombination at the constituent layer interfaces. This dissertation explores the possibility of scalable manufacturing of optoelectronics with low elastic modulus using perovskite polymer composites. Besides, this dissertation also studies the ion migration induced in-situ junction formation in halide perovskite polymer composite films and perovskite single crystals. Device failure mechanism caused by ion migration is also investigated in this dissertation. Perovskite thin film processing is essential for perovskite optoelectronics scalable manufacturing. In this dissertation, firstly, a uniform and pin-hole free thin film was processed using perovskite polymer composites. Perovskite LEDs were fabricated using the composite emitters. It has been discovered that an in situ homogeneous p-i-n junction can be developed in the composite emitter when an external bias is applied. The junction formation enables very efficient charge carrier transportation in perovskite LEDs without using additional electron transport layers (ETLs) and hole transport layers (HTLs). While a typical LED usually adopts a multi-layer structure, including both ETLs and HTLs. The unique simplified perovskite LED structure without using ETLs and HTLs is called "single-layer" structure. Moreover, scalable manufacturing of fully printed perovskite LEDs and intrinsically stretchable LEDs with robust mechanical performance is demonstrated benefiting from the "single-layer" structure in this dissertation. A stable junction formation is the basis of a stable perovskite LED. In this dissertation, the in-situ p-i-n homojunction in the perovskite polymer composites and perovskite single crystals are studied. AC impedance spectroscopy is used to study the junction formation and propagation of the perovskite polymer composites under an external electric field. Discharge current-voltage (I-V) characteristics and temperature dependence study are also conducted to support the ion migration induced junction formation and relaxation. It is a potential pathway to obtain highly stable perovskite LEDs by immobilizing the ions and stabilizing the junction. / A Dissertation submitted to the Department of Industrial and Manufacturing Engineering in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Fall Semester 2018. / October 31, 2018. / composite, junction formation, LEDs, perovskite, polymer / Includes bibliographical references. / Zhibin Yu, Professor Directing Dissertation; Jianping Zheng, University Representative; Zhiyong Liang, Committee Member; Omer Arda Vanli, Committee Member.

A microjet-based reactant delivery method was implemented in a prototype PEM fuel cell. A commercially available fuel cell was procured that utilizes a conventional reactant delivery method. The prototype was designed with an identical active area so that number of variables leading to differences in performance between the two fuel cells would be minimal. The MEAs inside each fuel cell were identical. Both the prototype and commercially available cell were tested under a variety of operating conditions. Conditions that were tested include: low air stoichiometry flow rates, high air stoichiometry flow rates, varying relative humidity, varying hydrogen stagnation pressure, varying stack temperatures, and varying air backpressure. The performance of each fuel cell was compared for each set of test conditions. The results show that the prototype microjet fuel cell experienced lower activation losses than the commercially available fuel cell. Testing revealed several design flaws in the prototype. Also, there are indications that the microjet design allows the fuel cell to be less affected by high stack temperatures. / A Thesis submitted to the Department of Mechanical Engineering in partial fulfillment of the requirements for the degree of Master of Science. / Degree Awarded: Summer Semester, 2008. / Date of Defense: June 10, 2008. / Fuel Cells, PEM / Includes bibliographical references. / Anjaneyulu Krothapalli, Professor Directing Thesis; Brenton Greska, Committee Member; Juan Ordonez, Committee Member.

Hurricane Andrew (1992) caused one of the largest property losses in U.S. history, but limited availability of surface wind measurements hindered the advancement of wind engineering research. Many studies have been conducted on regular boundary layer winds (non-hurricane winds) and their effects on the structures. In this case, their results were used in the standards and codes; however, hurricane winds and their effects on the structures still need more studies and observations. Analysis of hurricane surface winds revealed that turbulence spectrum of hurricane winds differs from that of non-hurricane surface winds. Vertical profile of wind velocity and turbulence intensity are also important for determining the wind loads on high-rise structures. Vertical profile of hurricane winds is affected by different parameters such as terrain or surface roughness. Recent studies show that wind velocity profile and turbulence intensity of hurricane winds may be different from those used in the design codes. Most of the studies and available models for analyzing wind turbines subjected to high-winds neglect unsteady aerodynamic forces on a parked wind tower. Since the blade pitch angle in a parked wind turbine is usually about 90°, the drag coefficient on blade airfoils are very small therefore the along-wind aerodynamic forces on the blades are smaller than those on the tower. Hence, the tower in parked condition plays an important role in along-wind responses of the wind turbine. The objectives of this study are, first, to explore the nature of the hurricane surface winds. Next, to establish a time domain procedure for addressing structure-wind-wave-soil interactions. Third, investigating the behavior of wind turbines subjected to hurricane loads resulted form hurricane nature and, lastly, to investigate reconfiguration of turbine structure to reduce wind forces. In order to achieve these objective, first, recent observations on hurricane turbulence models were discussed. Then a new formulation for addressing unsteady wind forces on the tower was introduced and NREL-FAST package was modified with the new formulation. Interaction of wind-wave-soil-structure was also included in the modification. After customizing the package, the tower and blade buffeting responses, the low cycle fatigue during different hurricane categories, and extreme value of the short-term responses were analyzed. In the second part, piezoelectric materials were used to generate perturbations on the surface of a specimen in the wind tunnel. This perturbation was used to combine upward wall motion and surface curvature. For this purpose, a Macro Fiber Composite (MFC) material was mounted on the surface of a cylindrical specimen for generating perturbation in the wind tunnel. Four different perturbation frequencies (1 Hz, 2 Hz, 3 Hz, and 4Hz) as well as the baseline specimen were tested in a low-speed wind tunnel (Re= 2.8×104). Results showed that recently observed turbulence models resulted in larger structural responses and low-cycle fatigue damage than existing models. In addition, extreme value analysis of the short-term results showed that the IEC 61400-3 recommendation for wind turbine class I was sufficient for designing the tower for wind turbine class S subjected to hurricane; however, for designing the blade, IEC 61400-3 recommendations for class I underestimated the responses. In addition, wind tunnel testing results showed that the perturbation of the surface of the specimen increased the turbulence in the leeward in specific distance from the specimen. The surface perturbation technique had potential to reduce the drag by 4.8%. / A Dissertation submitted to the Department of Civil and Environmental Engineering in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Fall Semester 2017. / October 31, 2017. / Includes bibliographical references. / Sungmoon Jung, Professor Directing Dissertation; Xiuwen Liu, University Representative; Lisa Spainhour, Committee Member; Michelle Rambo-Roddenberry, Committee Member.

Cavity structures, like weapons bays and landing gear wells on aircraft, suffer from severe oscillations under high speed flow conditions. These oscillations are associated with intense surface pressure/velocity fluctuations inside the cavity which can radiate strong acoustic waves and cause structural damage. The physics of cavity flows have been studied for several decades with much of the effort put towards flow controls to reduce these oscillations. Geometric modifications of the cavity structure are usually only effective for suppressing the oscillations within the designed flow conditions. Therefore, active flow control is more attractive for a wider application range. Previous research have proven that mass/momentum injection at the cavity leading edge can effectively suppress the pressure/velocity fluctuations. Due to the limited control authorities of current actuators, a steady actuation which introduces three-dimensional disturbances is studied to reduce the energy requirements of the actuator and improve the suppression of the oscillations over a wide range of free-stream Mach numbers. Surface fluctuating pressure measurements are acquired to determine the control performances of a number of 3-D actuation configurations. Flow fields, including velocity fields and density gradient fields, are measured to reveal the flow features with and without the flow control. Mathematical methods, including modal decomposition analysis, are further applied to study the dynamics of the flow field. All of these analyses together elucidate the effective 3-D actuation mechanism in the cavity flow control. The suppression of pressure fluctuations are obtained in both full-span and finite-span cavities. The successful flow control is found to be the redistribution of the energy in the shear layer by the counter-rotating-vortex pairs, which are introduced by the 3-D actuation in the cross-flow. In addition, a design guide for the actuator geometry is given based on the observations. / A Dissertation submitted to the Department of Mechanical Engineering in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Fall Semester 2017. / August 16, 2017. / Active flow control, Cavity flow, Steady blowing, Subsonic, Three-dimensional / Includes bibliographical references. / Louis N. Cattafesta, III, Professor Directing Dissertation; Christopher Tam, University Representative; Kunihiko Taira, Committee Member; Emmanuel G. Collins, Committee Member.